Choose your AI client below and follow the setup instructions.
You will need an API key. Sign in or create an account first, then return to this page to copy the right snippet for your client.
claude_desktop_config.json:{
"mcpServers": {
"uniclowd": {
"url": "https://uniclowd.com/api/chat/mcp",
"headers": {
"Authorization": "Bearer uc_YOUR_API_KEY"
}
}
}
}
Replace uc_YOUR_API_KEY with your API key from the dashboard.
Add UniClowd as a remote MCP server using the CLI:
claude mcp add uniclowd https://uniclowd.com/api/chat/mcp \
--header "Authorization: Bearer uc_YOUR_API_KEY"
Or add it to your project's .mcp.json:
{
"mcpServers": {
"uniclowd": {
"type": "url",
"url": "https://uniclowd.com/api/chat/mcp",
"headers": {
"Authorization": "Bearer uc_YOUR_API_KEY"
}
}
}
}
https://uniclowd.com/api/chat/openapi.json
Authorization: Bearer uc_YOUR_API_KEY.ChatGPT uses the REST API endpoints rather than the MCP streaming protocol.
{
"mcpServers": {
"uniclowd": {
"url": "https://uniclowd.com/api/chat/mcp",
"headers": {
"Authorization": "Bearer uc_YOUR_API_KEY"
}
}
}
}
Or add to your project's .cursor/mcp.json for project-level configuration.
Add UniClowd to your Gemini CLI MCP config at ~/.gemini/settings.json:
{
"mcpServers": {
"uniclowd": {
"uri": "https://uniclowd.com/api/chat/mcp",
"headers": {
"Authorization": "Bearer uc_YOUR_API_KEY"
}
}
}
}
Restart Gemini CLI after updating the config.
Use the UniClowd REST API with the xAI function calling format:
curl -X POST https://uniclowd.com/api/chat/rest/chat_send \
-H "Authorization: Bearer uc_YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"app": "telegram", "chat_id": "CHAT_ID", "text": "Hello"}'
Register each REST endpoint as a tool/function in your xAI agent. See the OpenAPI spec for all available endpoints.